Been making higher challenge monsters, and was noticing I had to really tune up their offenses compared to the existing monsters I've seen to feel like I was designing a suitable threat. That made me wonder what sort of benchmarks are built into the existing monsters as far as DPR per CR. I know damage isn't the only thing to consider, but it is A thing, and this is interesting...
I took the monster chart from Forge of Foes - they use a DPR (damage per round) which is mostly the average of the damage ranges given in the 2014 DMG page 274. It's worth noting that these values assume a monster is roughly equally balanced between defensive and offensive capability (at least whatever that means in game's maths).
I divided DPR by CR, organized the monster CRs by tier, leaving off the CR 21+ monsters for now, and here's what I found...
Tl;dr Average Damage per CR at Tier 0 (CR 1/8 to 1/2) is 21 damage... then at Tier 1 (CR 1 to CR 4) it drastically drops to about 8 damage... at Tier 2 (CR 5 to CR 10) it drops further to about 7 damage... at Tier 3 (CR 11 to CR 16) it drops further to about 6 damage... and at Tier 4 (CR 17 to CR 10) it stays about 6 damage with a minuscule drop. I'll post maths in spoilers.
This was fascinating to me, as my expectation would be the trend should go in the opposite direction – i.e. increasing damage per CR for monsters intended to challenge high-level characters.
Interesting find. So it is 6 more damage per round against 4 or 5 PCs that each usually get 5 to 10 hp per level.
I remember CR to be an indicator for a deadly monster, which means that it has a good chance to drop a PC in a single hit. That damage progression seems to achieve that against the average d6 class with 14 con.
Funny thing: just randomly checked a monster. Found adult green dragon at CR 15 which deals exactly 95 damage when using its legendary actions for their tail attack.
Honestly, I suspect this is on purpose: people get attached to their high-levels over time and don't want them to die.
They're running a game (or a narrative), not a simulator. They've tweaked the rules countless times and I'm sure something like this comes out in playtesting.
Mobility, action economy and legendary actions need to be considered for higher tier monsters too, not just average damage. Its the tactical use of abilities that makes a fun game not just increasing blam
This is quite interesting, but I think it's missing the other factor of offensive CR; namely, attack bonus. In 5e, AC is relatively static compared to earlier editions (particularly 3.x and 4e, both of which assumed a scaling AC), but attack bonus does scale with CR (it increases by +7 from CR 1/8 to CR 20).
If we look at it with attack bonus factored in, we get a significantly different result. I've based this on certain assumptions which may not be perfect. Consider this a rough draft meant to show my intent, rather than an analysis that is meant as the "be all end all". My math assumes that monsters will be attacking an average 17 AC, which does not improve with level. For attack bonus, I used the chart from DMG pg 274 (which, honestly, seems a bit low to me). I just expanded on your tables (it saved me from having to retype everything), I hope you don't mind.
CR 1/8 to 1/2
DPR
DPR per CR
ATT Bonus
Adjusted DPR
Adj DPR per CR
0.125
3
24
+3
1.05
8.4
0.25
5
20
+3
1.75
7
0.5
10
20
+3
3.5
7
AVERAGE
21.3333333333333
AVERAGE
7.47
CR 1 to 4
DPR
DPR per CR
ATT Bonus
Adjusted DPR
Adj DPR per CR
1
12
12
+3
4.2
4.2
2
17
8.5
+3
5.95
2.98
3
23
7.66666666666667
+4
9.2
3.07
4
29
7.25
+5
13.05
3.26
AVERAGE
8.85416666666667
AVERAGE
3.38
CR 5 to 10
DPR
DPR per CR
ATT Bonus
Adjusted DPR
Adj DPR per CR
5
35
7
+6
17.5
3.5
6
41
6.83333333333333
+6
20.5
3.42
7
47
6.71428571428571
+6
23.5
3.36
8
53
6.625
+7
29.15
3.64
9
59
6.55555555555556
+7
32.45
3.61
10
65
6.5
+7
35.75
3.58
AVG
6.70469576719577
AVERAGE
3.52
CR 11 to 16
DPR
DPR per CR
ATT Bonus
Adjusted DPR
Adj DPR per CR
11
71
6.45454545454545
+8
42.6
3.87
12
77
6.41666666666667
+8
46.2
3.85
13
83
6.38461538461538
+8
49.8
3.83
14
89
6.35714285714286
+8
53.4
3.81
15
95
6.33333333333333
+8
57
3.8
16
101
6.3125
+9
65.65
4.1
AVG
6.37646728271728
AVERAGE
3.88
CR 17 to 20
DPR
DPR per CR
ATT Bonus
Adjusted DPR
Adj DPR per CR
17
107
6.29411764705882
+10
74.9
4.4
18
113
6.27777777777778
+10
79.1
4.39
19
119
6.26315789473684
+10
83.3
4.38
20
132
6.6
+10
92.4
4.62
AVG
6.35876332989336
AVERAGE
4.45
From this we can see that, apart from CRs less than or equal to 1, monster's damage output slowly increases with CR over the tiers. Admittedly, this is based on my assumptions that PC ACs remain flat (on average) over the course of a campaign.
I would conclude that while average damage does appear to improve with CR, the scaling is unlikely to be sufficient to keep pace with the various mitigations characters gain access to as they level (uncanny dodge, CC spells, stunning strike, etc). The average adjusted improvement from CR 2 to CR 20 is +1.47 DRC per CR, which seems small overall.
I don't know what ya'll are expecting, but I don't know any character at level 20 that would easily shake off taking 132 points of damage. A DPR of 6 per CR is utterly meaningless in that context, giving a false impression that it's somehow less dangerous than the DPR per CR of 21 for a 1/8 CR creature, whose only going to deal 3 hp of damage on a hit. Without mitigation that CR 20 is going to drop any character with a d10 HD and Con of 13 or less in a single round's worth of attacks. The "average DPR per CR" calculation seems meaningless to me in relation to the pure DPR output (especially the sub CR 1 calculations), and most especially the chance of that blow connecting.
I'd be more inclined to look at a hit survivability calculation - CR DPR vs. character's average hit points at each level (factoring in character AC vs creature Attack bonus).
Because I'm lazy and didn't want to have to redo T1, I went with the following AC: T1=17, T2 = 18, T3 = 19, and T4 = 20. The results were interesting. I'll try to write up the tables tonight, when I have time.
To summarize, T1 remains 3.38 damage / CR. Unless you treat CR 1 as part of the fractional tier (which I'm beginning to think ought to be the case, as CR 1 is an outlier), in which case CRs 2-4 average 3.1.
T2 is 3.18.
T3 is 3.24.
T4 is 3.5.
Depending on how you look at it, DMG/CR does still rise across the tiers but by only a miniscule amount.
If you include CR 1 in T1, then damage actually dips in T2 and T3, but pulls slightly ahead in T4.
Either way, I think this demonstrates just how potent even seemingly insignificant AC boosts are in 5e.
I don't know what ya'll are expecting, but I don't know any character at level 20 that would easily shake off taking 132 points of damage. A DPR of 6 per CR is utterly meaningless in that context, giving a false impression that it's somehow less dangerous than the DPR per CR of 21 for a 1/8 CR creature, whose only going to deal 3 hp of damage on a hit. Without mitigation that CR 20 is going to drop any character with a d10 HD and Con of 13 or less in a single round's worth of attacks. The "average DPR per CR" calculation seems meaningless to me in relation to the pure DPR output (especially the sub CR 1 calculations), and most especially the chance of that blow connecting.
I'd be more inclined to look at a hit survivability calculation - CR DPR vs. character's average hit points at each level (factoring in character AC vs creature Attack bonus).
